Equivariant quantum circuits for learning on weighted graphs

Variational quantum algorithms are the leading candidate for advantage on near-term quantum hardware. When training a parametrized quantum circuit in this setting to solve a specific problem, the choice of ansatz is one of the most important factors that determines the trainability and performance of the algorithm. In quantum machine learning (QML), however, the literature on ansatzes that are motivated by the training data structure is scarce. In this work, we introduce an ansatz for learning tasks on weighted graphs that respects an important graph symmetry, namely equivariance under node permutations. We evaluate the performance of this ansatz on a complex learning task, namely neural combinatorial optimization, where a machine learning model is used to learn a heuristic for a combinatorial optimization problem. We analytically and numerically study the performance of our model, and our results strengthen the notion that symmetry-preserving ansatzes are a key to success in QML.

[1]  Phillip C. Lotshaw,et al.  Parameter Transfer for Quantum Approximate Optimization of Weighted MaxCut , 2022, ACM Transactions on Quantum Computing.

[2]  Patrick J. Coles,et al.  Group-Invariant Quantum Machine Learning , 2022, PRX Quantum.

[3]  A. Glos,et al.  Space-efficient binary optimization for variational quantum computing , 2022, npj Quantum Information.

[4]  Hongxiang Guo,et al.  Novel Architecture of Parameterized Quantum Circuit for Graph Convolutional Network , 2022, 2203.03251.

[5]  I. Ceylan,et al.  Equivariant Quantum Graph Circuits , 2021, ICML.

[6]  Patrick J. Coles,et al.  Generalization in quantum machine learning from few training data , 2021, Nature Communications.

[7]  Patrick J. Coles,et al.  Diagnosing Barren Plateaus with Tools from Quantum Optimal Control , 2021, Quantum.

[8]  Sofiène Jerbi,et al.  Quantum agents in the Gym: a variational quantum algorithm for deep Q-learning , 2021, Quantum.

[9]  P. McMahon,et al.  Expectation values from the single-layer quantum approximate optimization algorithm on Ising problems , 2020, Quantum Science and Technology.

[10]  Robert Koenig,et al.  Hybrid quantum-classical algorithms for approximate graph coloring , 2020, Quantum.

[11]  Nicos Christofides Worst-Case Analysis of a New Heuristic for the Travelling Salesman Problem , 1976, Operations Research Forum.

[12]  Jingbo B. Wang,et al.  Quantum Walk-Based Vehicle Routing Optimisation , 2021, Frontiers in Physics.

[13]  K. Kavukcuoglu,et al.  Highly accurate protein structure prediction for the human proteome , 2021, Nature.

[14]  Oriol Vinyals,et al.  Highly accurate protein structure prediction with AlphaFold , 2021, Nature.

[15]  Jin Zheng,et al.  Quantum Graph Convolutional Neural Networks , 2021, 2021 40th Chinese Control Conference (CCC).

[16]  Louis-Paul Henry,et al.  Quantum evolution kernel: Machine learning on graphs with programmable arrays of qubits , 2021, Physical Review A.

[17]  Joan Bruna,et al.  Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges , 2021, ArXiv.

[18]  Patrick J. Coles,et al.  Cost function dependent barren plateaus in shallow parametrized quantum circuits , 2021, Nature Communications.

[19]  Andrea Montanari,et al.  Learning with invariances in random features and kernel models , 2021, COLT.

[20]  Alec Radford,et al.  Zero-Shot Text-to-Image Generation , 2021, ICML.

[21]  Andrea Lodi,et al.  Combinatorial optimization and reasoning with graph neural networks , 2021, IJCAI.

[22]  Kunal Marwaha Local classical MAX-CUT algorithm outperforms p=2 QAOA on high-girth regular graphs , 2021, Quantum.

[23]  M. Cerezo,et al.  Variational quantum algorithms , 2020, Nature Reviews Physics.

[24]  M. Cerezo,et al.  Effect of barren plateaus on gradient-free optimization , 2020, Quantum.

[25]  Nathan Wiebe,et al.  Entanglement Induced Barren Plateaus , 2020, PRX Quantum.

[26]  Patrick J. Coles,et al.  Noise-induced barren plateaus in variational quantum algorithms , 2020, Nature Communications.

[27]  Yoshua Bengio,et al.  Machine Learning for Combinatorial Optimization: a Methodological Tour d'Horizon , 2018, Eur. J. Oper. Res..

[28]  Jianwu Long,et al.  A Graph Neural Network for superpixel image classification , 2021 .

[29]  Owen Lockwood,et al.  Reinforcement Learning with Quantum Variational Circuit , 2020, AIIDE.

[30]  Adam Glos,et al.  Space-efficient binary optimization for variational computing , 2020, 2009.07309.

[31]  Quoc V. Le,et al.  Chip Placement with Deep Reinforcement Learning , 2020, ArXiv.

[32]  David Gamarnik,et al.  The Quantum Approximate Optimization Algorithm Needs to See the Whole Graph: A Typical Case , 2020, ArXiv.

[33]  Robert Koenig,et al.  Obstacles to State Preparation and Variational Optimization from Symmetry Protection. , 2019, 1910.08980.

[34]  Jacob Biamonte,et al.  On the universality of the quantum approximate optimization algorithm , 2019, Quantum Information Processing.

[35]  Chao-Han Huck Yang,et al.  Variational Quantum Circuits for Deep Reinforcement Learning , 2019, IEEE Access.

[36]  Zhiyuan Liu,et al.  Graph Neural Networks: A Review of Methods and Applications , 2018, AI Open.

[37]  Leo Zhou,et al.  Quantum Approximate Optimization Algorithm: Performance, Mechanism, and Implementation on Near-Term Devices , 2018, Physical Review X.

[38]  Mario Szegedy,et al.  What do QAOA energies reveal about graphs , 2019, 1912.12277.

[39]  Robin M. Schmidt Recurrent Neural Networks (RNNs): A gentle Introduction and Overview , 2019, ArXiv.

[40]  Jack Hidary,et al.  Quantum Graph Neural Networks , 2019, ArXiv.

[41]  Marcello Benedetti,et al.  Parameterized quantum circuits as machine learning models , 2019, Quantum Science and Technology.

[42]  Elman Mansimov,et al.  Molecular Geometry Prediction using a Deep Generative Graph Neural Network , 2019, Scientific Reports.

[43]  Yuan He,et al.  Graph Neural Networks for Social Recommendation , 2019, WWW.

[44]  Soonwon Choi,et al.  Quantum convolutional neural networks , 2018, Nature Physics.

[45]  Philip S. Yu,et al.  A Comprehensive Survey on Graph Neural Networks , 2019, IEEE Transactions on Neural Networks and Learning Systems.

[46]  Harper R. Grimsley,et al.  ADAPT-VQE: An exact variational algorithm for fermionic simulations on a quantum computer , 2018, 1812.11173.

[47]  S. Lloyd Quantum approximate optimization is computationally universal , 2018, 1812.11075.

[48]  F. Brandão,et al.  For Fixed Control Parameters the Quantum Approximate Optimization Algorithm's Objective Function Value Concentrates for Typical Instances , 2018, 1812.04170.

[49]  H. Neven,et al.  Barren plateaus in quantum neural network training landscapes , 2018, Nature Communications.

[50]  Lawrence V. Snyder,et al.  Reinforcement Learning for Solving the Vehicle Routing Problem , 2018, NeurIPS.

[51]  J. Gambetta,et al.  Hardware-efficient variational quantum eigensolver for small molecules and quantum magnets , 2017, Nature.

[52]  Le Song,et al.  2 Common Formulation for Greedy Algorithms on Graphs , 2018 .

[53]  A. Harrow,et al.  Quantum Supremacy through the Quantum Approximate Optimization Algorithm , 2016, 1602.07674.

[54]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[55]  G. Harman,et al.  $$\sin (\omega x)$$sin(ωx) Can Approximate Almost Every Finite Set of Samples , 2015 .

[56]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[57]  E. Farhi,et al.  A Quantum Approximate Optimization Algorithm , 2014, 1411.4028.

[58]  Alán Aspuru-Guzik,et al.  A variational eigenvalue solver on a photonic quantum processor , 2013, Nature Communications.

[59]  Andrew Lucas,et al.  Ising formulations of many NP problems , 2013, Front. Physics.

[60]  Ameet Talwalkar,et al.  Foundations of Machine Learning , 2012, Adaptive computation and machine learning.

[61]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[62]  Yoshua Bengio,et al.  Convolutional networks for images, speech, and time series , 1998 .

[63]  Gerd Hirzinger,et al.  Why feed-forward networks are in a bad shape , 1998 .